Tải bản đầy đủ
4 Goldratt’s resistance model 239

4 Goldratt’s resistance model 239

Tải bản đầy đủ

240

Critical Chain Project Management

longer than their (resource-unleveled) critical path plan and beyond
management’s due date before they add the project buffer. That leads to
consternation and often stops progress.
The project planners put themselves into the pickle. I caution them in
training that people do not know the real difference between a 50-50 estimate and a low-risk estimate, which experience has demonstrated many
times. For that reason, people will give you an estimate when asked and
characterize it as the type of estimate you asked for. If you start by asking
for 50-50 estimates, people will believe that is what they gave you. You
will now have a difficult time getting their agreement to a plan with
shorter duration than the ones they have framed in their minds as 50-50
estimates. This is a case where you really do go all the way back to layer 1
of resistance.
People often bring up fears about potential unintended consequences
of implementing critical chain as a reason not to proceed. I encourage
them to consider such potentials and to then design a way that prevents
or mitigates the unintended consequence. However, experience demonstrates that it is not a good idea to plan to prevent all potential occurrences
of unintended consequence before you start. As President Eisenhower
said, “Nothing would ever be attempted if all objections had to first be
overcome.”
For example, a frequent concern is that if you succeed to implement
multiproject critical chain, some customers will receive more of the benefit of project acceleration than others will. Even though all the projects
will get done quicker than any did before, some people are concerned that
less acceleration will cause some customers to be unhappy. I originally
marveled at expressing that concern, because the only implied solution is
to not implement critical chain, that is, to ensure that all customers get
their results equally late. Of course, there are other solutions.
You might now ask, “Are there ways I can make use of my newfound
ability to deliver early and satisfy all my customers?” Most people, after
getting over the shock of having to respond positively to a negative
concern they raised, are able to think up several ways. For example,
offering acceleration at a premium price or simply beginning to deliver
early against normal lead times. Such strategies amount to market
segmentation.
Some customers judge responsiveness by when you start projects
rather than by when you finish. Contractors to the U.S. government most

Implementing the change to critical chain

241

often express those concerns. In such cases, the client gets involved in
many details of project performance. Many contracts, for example,
require data submissions x days after the contract has been signed. Once
again, having surfaced, this concern gives you the information you need
to prevent or mitigate it from being a problem for you or your customer. It
is never a reason to forego the benefits offered by CCPM.
Although I always deal with the issue in two-day training sessions,
layer 4 usually comes out again during implementation as some form of
“What if (management, the client) cuts my buffer, takes away my buffer,
or simply judges my success relative to the start of the buffer?” Successful
implementations have never had that problem in reality, but explaining
that to people standing on the precipice does not bring sufficient peace
of mind to unlock their paradigm and cause them to leap.

9.4.3

Overcoming layer 5

The most common manifestation of layer 5 is that management will not
change. I dealt with this with a client as I was writing this paragraph.
Management has a system of quality gates, or milestones, that must be
achieved to complete their projects. They also have a reporting system to
inform management of progress to those milestones. The reporting
system requires specific dates for each milestone and reporting to each
date once a month. The response to my suggestion to take this to management and point out the inconsistency with the critical chain paradigm
was met with, “I am not going to suggest to them that they change their
measurements. They won’t listen to me.” That person is now on a path to
overcome that obstacle, successfully I am sure.
Equally often, it is the customer who will not change. Or it may be
the regulators. Often, it is other organizational elements. The evaporating cloud is a useful tool to overcome the obstacles that become real
during implementation. Consider going back to President Eisenhower’s
view—nothing would ever be accomplished if all objections had to be
overcome first.

9.4.4

Overcoming layer 6

Some have suggested that the first five layers of resistance are the active
phase, in which people feel safe to raise objections. After all of the

242

Critical Chain Project Management

objections have been answered, they have no recourse but to go into passive resistance. That is, they seem to support the idea, they just do not do
it. People use what Argyris called “skilled incompetence” to avoid change
(which he called “organizational learning”) [6]. For example, they make
certain things are implicitly undiscussable in the organization and then cover
it up by making its undiscussability undiscussable. Argyris reports that his
research shows nearly all people hold these theories-in-use and that:
Individuals may unknowingly provide us with distorted information,
and those same individuals may hesitate to engage in the dialogue that
is required to explore the possibility of such distortions. If we persist
in exploring these issues, practitioners may become defensive—this
defensiveness leading, in turn, to new distortions, both recognized and
unrecognized. [6]

The most effective means of combating layer-6 resistances is to just
move ahead and listen a lot. When leadership follows the implementation plan and expects others to follow it, following the CCPM behavior
patterns has not been difficult. You have to start the positive feedback
loop of project success. The feedback loop will sustain the projects, and
that will sustain the implementation.

9.5

To pilot or not to pilot?

The second most common response offered by organizations considering
changing to critical chain is, “Let’s try a pilot project.” (The most common
is, “Can you show evidence of companies just like us that have tried this
and succeeded?” That from people who want to become leaders of their
industry, just as long as they follow everyone else. Go figure.) They
hope that the pilot project will help them reduce the risk of full-scale
implementation.
Figure 9.9 illustrates the pilot project evaporating cloud. The upper
branch supports the desire for the pilot project, while the lower branch
supports not doing the pilot project. Some of the assumptions under the
arrows of this cloud include:
◗ B-to-A arrow: There may be something unique about our environ-

ment that invalidates the theory and the experience of others.

Implementing the change to critical chain
B
Assure that the
method will work in
our system before
risking full-scale
implementation

243

D
Perform pilot project

A
Successful system
improvement
C
Achieve all of the
improvement benfits,
as soon as possible,
on as many projects
as possible

Figure 9.9

D′
Do not perform
pilot project

The pilot project dilemma.

◗ D-to-B arrow: We cannot anticipate or work our way through the

things that might block our implementation or cause unintended
consequences.
◗ D’-to-C arrow: The pilot project does not allow for the multiproject

benefits.
◗ D’-to-C arrow: The pilot project will delay the benefits to other

projects.
◗ D’-to-C arrow: The pilot project will not cause behavior change for

resources supporting both the pilot project (with CCPM) and other
projects using past practice.
Pilot projects sometimes succeed. They can succeed for several reasons, including that the immediately preceding projects were such dismal
failures that sheer chance (or regression to the mean, for the statisticians
among you) makes it likely that the next project will succeed, no matter
what you change or do not change. In addition, you may witness
the Hawthorne effect, named after the Western Electric Company’s
Hawthorne Works, near Chicago, where some early scientific management experiments took place. It describes the response of people broken
out and given special treatment. The experiments changed things about
production. Whatever they changed seemed to improve performance.